Background: Diffusion-weighted imaging radiomics could be used as prognostic biomarkers in acute ischemic stroke. We aimed to identify a clinical and diffusion-weighted imaging radiomics model for individual unfavorable outcomes risk assessment in acute ischemic stroke. Methods: A total of 1716 patients with acute ischemic stroke from 2 centers were divided into a training cohort and a validation cohort. Patient outcomes were measured with the modified Rankin Scale score. An unfavorable outcome was defined as a modified Rankin Scale score greater than 2. The primary end point was all-cause mortality or outcomes 1 year after stroke. The MRI-DRAGON score was calculated based on previous publications. We extracted and selected the infarct features on diffusion-weighted imaging to construct a radiomic signature. The clinic-radiomics signature was built by measuring the Cox proportional risk regression score (CrrScore) and compared with the MRI-DRAGON score and the ClinicScore. CrrScore model performance was estimated by 1-year unfavorable outcomes prediction. Results: A high radiomic signature predicted a higher probability of unfavorable outcomes than a low radiomic signature in the training (hazard ratio, 3.19 [95% CI, 2.51–4.05]; P <0.0001) and validation (hazard ratio, 3.25 [95% CI, 2.20–4.80]; P <0.0001) cohorts. The diffusion-weighted imaging Alberta Stroke Program Early CT Score, age, glucose level before therapy, National Institutes of Health Stroke Scale score on admission, glycated hemoglobin‚ radiomic signature, hemorrhagic infarction, and malignant cerebral edema were associated with an unfavorable outcomes risk after multivariable adjustment. A CrrScore nomogram was developed to predict outcomes and had the best performance in the training (area under the curve, 0.862) and validation cohorts (area under the curve, 0.858). The CrrScore model time-dependent areas under the curve of the probability of unfavorable outcomes at 1 year in the training and validation cohorts were 0.811 and 0.801, respectively. Conclusions: The CrrScore model allows the accurate prediction of patients with acute ischemic stroke outcomes and can potentially guide rehabilitation therapies for patients with different risks of unfavorable outcomes.

Background: Accurately distinguishing between Parkinson disease (PD) and healthy controls (HCs) through reliable imaging method is crucial for appropriate therapeutic intervention. However, PD diagnosis is hindered by the subjective nature of the evaluation. We aimed to develop an automatic deep-learning method that can segment the substantia nigra areas on susceptibility-weighted imaging (SWI) and T2-weighted imaging (T2WI) and further differentiate patients with PD from HCs using a machine learning algorithm. Methods: Magnetic resonance imaging (MRI) data from 83 patients with PD and 83 age- and sex-matched HCs were obtained on the same 3.0-T MRI scanner. A deep learning method with Swin-Unet was developed to segment volumes of interest (VOIs) on SWI and then map the VOIs on SWI to the corresponding T2WI; features were then extracted from the VOIs on SWI and T2WI. Three machine learning models were developed and compared to differentiate those with PD from HCs. Results: Swin-Unet achieved a better Dice coefficient than did U-Net in SWI segmentation (0.832 vs. 0.712). Machine learning models outperformed visual analysis (P>0.05), and logistic regression (LR) achieved the best performance [area under the curve (AUC) ≥0.819] and the most stable (relative standard deviations in AUC ≤0.05). The test results showed that the AUC of the LR model based on SWI segmentation was 0.894 while that of the LR model based on T2WI segmentation was 0.876. There was no significant difference in VOIs based on manual labeling or automatic segmentation across T2WI, SWI, or a combination of the two (P>0.05). The AUCs of the LR model based on automatic segmentation were close to those of the model based on manual labeling (P>0.05). Conclusions: Our approach could provide a powerful and useful method for automatically and rapidly diagnosing PD in the clinic with only T2WI.

Glioma grading plays an important role in surgical resection. We investigated the ability of different feature reduction methods in support vector machine (SVM)-based diffusion kurtosis imaging (DKI) histogram parameters to distinguish glioma grades. A total of 161 glioma patients who underwent magnetic resonance imaging (MRI) from January 2017 to January 2020 were included retrospectively. The patients were divided into low-grade (n = 61) and high-grade (n = 100) groups. Parametric DKI maps were derived, and 45 features from the DKI maps were extracted semi-automatically for analysis. Three feature selection methods [principal component analysis (PCA), recursive feature elimination (RFE) and least absolute shrinkage and selection operator (LASSO)] were used to establish the glioma grading model with an SVM classifier. To evaluate the performance of SVM models, the receiver operating characteristic (ROC) curves of SVM models for distinguishing glioma grades were compared with those of conventional statistical methods. The conventional ROC analysis showed that mean diffusivity (MD) variance, MD skewness and mean kurtosis (MK) C50 could effectively distinguish glioma grades, particularly MD variance. The highest classification distinguishing AUC was found using LASSO at 0.904 ± 0.069. In comparison, classification AUC by PCA was 0.866 ± 0.061, and 0.899 ± 0.079 by RFE. The SVM-PCA model with the lowest AUC among the SVM models was significantly better than the conventional ROC analysis (z = 1.947, p = 0.013). These findings demonstrate the superiority of DKI histogram parameters by LASSO analysis and SVM for distinguishing glioma grades.

Background: The segmentation of cerebral vasculature on magnetic resonance angiography (MRA) images is crucial for both clinical applications and research. However, existing cerebrovascular segmentation techniques fail to account for the complex geometric and topological features of cerebral vasculature, resulting in suboptimal segmentation of distal small arteries. This study aimed to develop a novel deep vascular network (DVNet) with dual contextual path (DCP) and vascular attention enhancement module (VAEM) to accurately segment the cerebrovasculature. Methods: First, we developed DVNet with DCP and VAEM to accurately segment the cerebral vasculature from the publicly available dataset MR Brain Images of Healthy Volunteers (MIDAS)-I. Second, the segmentation performance of the proposed DVNet was evaluated using the MIDAS-II dataset, followed by further validation using our hospital data. Finally, ablation experiments were performed to evaluate the segmentation performance. Results: Experiments show that the proposed segmentation approach outperforms the previously proposed approaches (U-Net, V-Net, endoplasmic reticulum Net, etc.), with 0.900 Dice and 0.860 mean Intersection over Union (IoU) in the MIDAS-I dataset and 0.715 Dice and 0.713 IoU in the MIDAS-II dataset. External validation revealed good segmentation performance (Dice coefficient: 0.733; IoU: 0.730). Conclusions: Our approach demonstrates that accurate and robust cerebrovascular segmentation is achievable on MRA using DVNet with DCP and VAEM, especially in small distal vessels.

The early prediction of malignant cerebral edema (MCE) following endovascular therapy for acute ischemic stroke is of paramount importance for facilitating timely interventions. The present study aimed to create a comprehensive map of lesion topography associated with MCE risk and to build a machine learning model based on these topography-informed radiomics to predict the MCE in stroke patients after endovascular therapy. Using voxel-based lesion analyses, we comprehensively quantified the spatial features of infarct location lesions. These topological features were integrated with radiomics to create a hybrid spatial radiomics model. Four machine learning algorithms bases on topography features, radiomics, and Topo-Rad features were developed to predict MCE in acute stroke patients, respectively. The performance of models was evaluated using the receiver operating characteristic curves, decision curve analysis and Net Reclassification Improvement. The SHapley Additive exPlanations (SHAP) method was employed to interpret and visualize the output of the optimal model. The topography maps for acute stroke patients showed the right temporal lobe and right caudate nucleus were significantly associated with MCE (P < 0.05). For four ML algorithms, the SVM model based on topo-Rad achieved the highest predictive performance (AUC in training/validation set: 0.872/0.842), while no statistically significant difference was observed compared to the model based on topography (0.857/0.812). The SHAP plots demonstrated that the most significant contributors to model performance were related to temporal_pars_of_MCA_R, occipital_pars_of_PCA_R, parietal_pars_of_MCA_R, temporal_pars_of_MCA_L, and parietal_pars_of_MCA_L. The infarct topography plays a dominant role in predicting MCE following endovascular therapy, with radiomic features providing limited additional predictive value.